arXiv subjects

Yikun Cheng

Publications and source records attributed to Yikun Cheng.

11 recordsLinked to original sources

arXiv2026

Neural-NPV Control: Learning Parameter-Dependent Controllers and Lyapunov Functions

This paper presents Neural-NPV Control, a learning-based framework for joint synthesis of a parameter-dependent (PD) controller and a PD Lyapunov function using neural networks for an NPV system under input constraints. At the first stage, the proposed framework utilizes a gradient-based counterexample-guided procedure to synthesize a PD controller and a PD Lyapunov function candidate. The second stage relies on a level-set guided procedure to refine the controller and Lyapunov function candidate while maximizing the robust region of attraction (R-ROA). The learned controller, Lyapunov function, and R-ROA are empirically evaluated. We demonstrate the advantages of Neural-NPV over SOS-based methods in terms of applicability, performance, and scalability through numerical experiments involving a simple inverted pendulum with one scheduling parameter and a quadrotor system with three scheduling parameters.

eess.SY↗

arXiv2026

A Deployable Bio-inspired Compliant Leg Design for Enhanced Leaping in Quadruped Robots

Quadruped robots are becoming increasingly essential for various applications, including industrial inspection and catastrophe search and rescue. These scenarios require robots to possess enhanced agility and obstacle-navigation skills. Nonetheless, the performance of current platforms is often constrained by insufficient peak motor power, limiting their ability to perform explosive jumps. To address this challenge, this paper proposes a bio-inspired method that emulates the energy-storage mechanism found in froghopper legs. We designed a Deployable Compliant Leg (DCL) utilizing a specialized 3D-printed elastic material, Polyether block amide (PEBA), featuring a lightweight internal lattice structure. This structure functions analogously to biological tendons, storing elastic energy during the robot's squatting phase and rapidly releasing it to augment motor output during the leap. The proposed mechanical design significantly enhances the robot's vertical jumping capability. Through finite element analysis (FEA) and experimental validation, we demonstrate a relative performance improvement of 17.1% in vertical jumping height.

cs.RO↗

arXiv2025

Needle Biopsy And Fiber-Optic Compatible Robotic Insertion Platform

Tissue biopsy is the gold standard for diagnosing many diseases, involving the extraction of diseased tissue for histopathology analysis by expert pathologists. However, this procedure has two main limitations: 1) Manual sampling through tissue biopsy is prone to inaccuracies; 2) The extraction process is followed by a time-consuming pathology test. To address these limitations, we present a compact, accurate, and maneuverable robotic insertion platform to overcome the limitations in traditional histopathology. Our platform is capable of steering a variety of tools with different sizes, including needle for tissue extraction and optical fibers for vibrational spectroscopy applications. This system facilitates the guidance of end-effector to the tissue and assists surgeons in navigating to the biopsy target area for multi-modal diagnosis. In this paper, we outline the general concept of our device, followed by a detailed description of its mechanical design and control scheme. We conclude with the validation of the system through a series of tests, including positioning accuracy, admittance performance, and tool insertion efficacy.

cs.RO↗

arXiv2025

DBaS-Log-MPPI: Efficient and Safe Trajectory Optimization via Barrier States

Optimizing trajectory costs for nonlinear control systems remains a significant challenge. Model Predictive Control (MPC), particularly sampling-based approaches such as the Model Predictive Path Integral (MPPI) method, has recently demonstrated considerable success by leveraging parallel computing to efficiently evaluate numerous trajectories. However, MPPI often struggles to balance safe navigation in constrained environments with effective exploration in open spaces, leading to infeasibility in cluttered conditions. To address these limitations, we propose DBaS-Log-MPPI, a novel algorithm that integrates Discrete Barrier States (DBaS) to ensure safety while enabling adaptive exploration with enhanced feasibility. Our method is efficiently validated through three simulation missions and one real-world experiment, involving a 2D quadrotor and a ground vehicle navigating through cluttered obstacles. We demonstrate that our algorithm surpasses both Vanilla MPPI and Log-MPPI, achieving higher success rates, lower tracking errors, and a conservative average speed.

cs.RO↗

arXiv2025

MPPI-DBaS: Safe Trajectory Optimization with Adaptive Exploration

In trajectory optimization, Model Predictive Path Integral (MPPI) control is a sampling-based Model Predictive Control (MPC) framework that generates optimal inputs by efficiently simulating numerous trajectories. In practice, however, MPPI often struggles to guarantee safety assurance and balance efficient sampling in open spaces with the need for more extensive exploration under tight constraints. To address this challenge, we incorporate discrete barrier states (DBaS) into MPPI and propose a novel MPPI-DBaS algorithm that ensures system safety and enables adaptive exploration across diverse scenarios. We evaluate our method in simulation experiments where the vehicle navigates through closely placed obstacles. The results demonstrate that the proposed algorithm significantly outperforms standard MPPI, achieving a higher success rate and lower tracking errors.

eess.SY↗

arXiv2024

CROPS: A Deployable Crop Management System Over All Possible State Availabilities

Exploring the optimal management strategy for nitrogen and irrigation has a significant impact on crop yield, economic profit, and the environment. To tackle this optimization challenge, this paper introduces a deployable \textbf{CR}op Management system \textbf{O}ver all \textbf{P}ossible \textbf{S}tate availabilities (CROPS). CROPS employs a language model (LM) as a reinforcement learning (RL) agent to explore optimal management strategies within the Decision Support System for Agrotechnology Transfer (DSSAT) crop simulations. A distinguishing feature of this system is that the states used for decision-making are partially observed through random masking. Consequently, the RL agent is tasked with two primary objectives: optimizing management policies and inferring masked states. This approach significantly enhances the RL agent's robustness and adaptability across various real-world agricultural scenarios. Extensive experiments on maize crops in Florida, USA, and Zaragoza, Spain, validate the effectiveness of CROPS. Not only did CROPS achieve State-of-the-Art (SOTA) results across various evaluation metrics such as production, profit, and sustainability, but the trained management policies are also immediately deployable in over of ten millions of real-world contexts. Furthermore, the pre-trained policies possess a noise resilience property, which enables them to minimize potential sensor biases, ensuring robustness and generalizability. Finally, unlike previous methods, the strength of CROPS lies in its unified and elegant structure, which eliminates the need for pre-defined states or multi-stage training. These advancements highlight the potential of CROPS in revolutionizing agricultural practices.

cs.AI↗

arXiv2024

Transformer Based Tissue Classification in Robotic Needle Biopsy

Image-guided minimally invasive robotic surgery is commonly employed for tasks such as needle biopsies or localized therapies. However, the nonlinear deformation of various tissue types presents difficulties for surgeons in achieving precise needle tip placement, particularly when relying on low-fidelity biopsy imaging systems. In this paper, we introduce a method to classify needle biopsy interventions and identify tissue types based on a comprehensive needle-tissue contact model that incorporates both position and force parameters. We trained a transformer model using a comprehensive dataset collected from a formerly developed robotics platform, which consists of synthetic and porcine tissue from various locations (liver, kidney, heart, belly, hock) marked with interaction phases (pre-puncture, puncture, post-puncture, neutral). This model achieves a significant classification accuracy of 0.93. Our demonstrated method can assist surgeons in identifying transitions to different tissues, aiding surgeons with tissue awareness.

eess.SP↗

arXiv2022

Safe and Efficient Reinforcement Learning Using Disturbance-Observer-Based Control Barrier Functions

Safe reinforcement learning (RL) with assured satisfaction of hard state constraints during training has recently received a lot of attention. Safety filters, e.g., based on control barrier functions (CBFs), provide a promising way for safe RL via modifying the unsafe actions of an RL agent on the fly. Existing safety filter-based approaches typically involve learning of uncertain dynamics and quantifying the learned model error, which leads to conservative filters before a large amount of data is collected to learn a good model, thereby preventing efficient exploration. This paper presents a method for safe and efficient RL using disturbance observers (DOBs) and control barrier functions (CBFs). Unlike most existing safe RL methods that deal with hard state constraints, our method does not involve model learning, and leverages DOBs to accurately estimate the pointwise value of the uncertainty, which is then incorporated into a robust CBF condition to generate safe actions. The DOB-based CBF can be used as a safety filter with model-free RL algorithms by minimally modifying the actions of an RL agent whenever necessary to ensure safety throughout the learning process. Simulation results on a unicycle and a 2D quadrotor demonstrate that the proposed method outperforms a state-of-the-art safe RL algorithm using CBFs and Gaussian processes-based model learning, in terms of safety violation rate, and sample and computational efficiency.

cs.RO↗

arXiv2022

Robustifying Reinforcement Learning Policies with $\mathcal{L}_1$ Adaptive Control

A reinforcement learning (RL) policy trained in a nominal environment could fail in a new/perturbed environment due to the existence of dynamic variations. Existing robust methods try to obtain a fixed policy for all envisioned dynamic variation scenarios through robust or adversarial training. These methods could lead to conservative performance due to emphasis on the worst case, and often involve tedious modifications to the training environment. We propose an approach to robustifying a pre-trained non-robust RL policy with $\mathcal{L}_1$ adaptive control. Leveraging the capability of an $\mathcal{L}_1$ control law in the fast estimation of and active compensation for dynamic variations, our approach can significantly improve the robustness of an RL policy trained in a standard (i.e., non-robust) way, either in a simulator or in the real world. Numerical experiments are provided to validate the efficacy of the proposed approach.

cs.LG↗

arXiv2021

Guaranteed Trajectory Tracking under Learned Dynamics with Contraction Metrics and Disturbance Estimation

This paper presents an approach to trajectory-centric learning control based on contraction metrics and disturbance estimation for nonlinear systems subject to matched uncertainties. The approach uses deep neural networks to learn uncertain dynamics while still providing guarantees of transient tracking performance throughout the learning phase. Within the proposed approach, a disturbance estimation law is adopted to estimate the pointwise value of the uncertainty, with pre-computable estimation error bounds (EEBs). The learned dynamics, the estimated disturbances, and the EEBs are then incorporated in a robust Riemann energy condition to compute the control law that guarantees exponential convergence of actual trajectories to desired ones throughout the learning phase, even when the learned model is poor. On the other hand, with improved accuracy, the learned model can help improve the robustness of the tracking controller, e.g., against input delays, and can be incorporated to plan better trajectories with improved performance, e.g., lower energy consumption and shorter travel time.The proposed framework is validated on a planar quadrotor example.

eess.SY↗

arXiv2020

Safety Constrained Multi-UAV Time Coordination: A Bi-level Control Framework in GPS Denied Environment

Unmanned aerial vehicles (UAVs) suffer from sensor drifts in GPS denied environments, which can cause safety issues. To avoid intolerable sensor drifts while completing the time-critical coordination task for multi-UAV systems, we propose a safety constrained bi-level control framework. The first level is the time-critical coordination level that achieves a consensus of coordination states and provides a virtual target which is a function of the coordination state. The second level is the safety-critical control level that is designed to follow the virtual target while adapting the attacked UAV(s) at a path re-planning level to support resilient state estimation. In particular, the time-critical coordination level framework generates the desired speed and position profile of the virtual target based on the multi-UAV cooperative mission by the proposed consensus protocol algorithm. The safety-critical control level is able to make each UAV follow its assigned path while detecting the attacks, estimating the state resiliently, and driving the UAV(s) outside the effective range of the spoofing device within the escape time. The numerical simulations of a three-UAV system demonstrate the effectiveness of the proposed safety constrained bi-level control framework.

eess.SY↗